Precambrian Research 101 (2000) 263–275

The relation between iron-formation and low temperaturehydrothermal alteration in an Archean volcanic environment

E.H. Chown *, E. N’dah, W.U. MuellerSciences de la Terre, Uni6ersite du Quebec a Chicoutimi, Chicoutimi, Quebec, Canada G7H 2B1

Abstract

The uppermost section of the Hunter Mine group (HMG) (2728 Ma), a bimodal volcanic complex in the Abitibigreenstone belt, contains both oxide and carbonate facies banded iron-formation (BIF). This paper explores therelationship between volcanic activity and the development of the two types of iron-formation. The oxide facies,represented by chert-jasper-magnetite iron-formation is widespread and recurs at several stratigraphic levels. Fieldevidence suggests it deposited directly on the sea floor during periods of volcanic quiescence, and is most probablyderived from fluids seeping from volcaniclastic sediments and the compaction of felsic shard-rich tuffs. Seepingoccurred along synsedimentary and synvolcanic faults and then along bedding planes. Carbonate (siderite) faciesiron-formation, in contrast, formed locally below a silica cap rock by in situ low-temperature hydrothermalreplacement of chert-tuff beds and even of the oxide formation. Thus the two types of iron-formation, althoughspatially and stratigraphically juxtaposed, are not the result of the same process, and only the carbonate faciesappears to be directly related to a low temperature hydrothermal volcanic process. The contrasting genesis of oxideand carbonate iron-formations indicates that current simplistic models relating banded iron-formation to volcanicmassive sulphide deposits need to be re-evaluated. The notion that oxide iron-formation represents a distalmanifestation of volcano-generated hydrothermal activity may be the exception and not the rule, because, as indicatedby this study, BIF occurs in the central part of a volcanic edifice, rather than on the flanks. © 2000 Elsevier ScienceB.V. All rights reserved.

Keywords: Iron-formation; Oxide; Carbonate; Low temperature; Hydrothermal; Volcanic; Archean

www.elsevier.com/locate/precamres

1. Introduction

The origins of iron-formations remain contro-versial (Kimberley, 1979, 1989; Dimroth, 1986).Banded iron-formations (BIF), prominent duringthe Archean-Paleoproterozoic, have been used to

indicate reducing atmospheric conditions duringEarth’s evolution (Holland, 1984). The most re-cent classifications of iron-formations, or chert-bearing ironstones, (Kimberley, 1979, 1989; Fyonet al., 1992) distinguish those of shallow- anddeep-water settings (Simonson, 1985), as well asdirectly deposited and replacement types of iron-formations. In Archean volcanic arc settings Fyonet al. (1992) recognize a shallow-water type whichmay deposit on eroded volcanic edifices, and the

* Corresponding author. Present address: 90 Dickens Drive,Kingston, Ont., Canada K7M 2M8.

E-mail address: madned@kingston.net (E.H. Chown)

0301-9268/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.

PII: S 0301 -9268 (99 )00091 -1

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275264

more common deep-water type with a proximalsulphide-graphitic shale facies and the more com-mon distal oxide-turbidite facies. Lithologic asso-ciations may determine the depositional setting ofan iron-formation, but the possible replacementorigin may be difficult to determine in metamor-phosed sequences.

James (1954) first documented facies associa-tions (oxide, silicate and carbonate) in shallow-water iron-formations, attributing the changes towater depth at the time of deposition. However,subsequent work has indicated that the mineral-ogy may be early diagenetic (Dimroth and Chau-vel, 1973; Kimberley, 1979). The source of thediagenetic fluid is more contentious but a primaryor secondary volcanic source is generally indi-cated (Dimroth, 1986). Deep-water banded iron-formations with lutite interbeds are less wellunderstood, but their formation, by analogy withthat of the shallow-water type, is generally at-tributed to basinal exhalation and subsequentchemical sedimentation. The Fe- and Si-rich fluidsare either of direct volcanic origin (Gross, 1980;Lydon, 1984) or deep circulating waters enrichedby water-rock reactions (Kimberley, 1994). Facies

variation appears to be related to source proxim-ity (Simonson, 1985; Gross, 1980). Gross (1980)suggested BIF’s were products of volcanic exhala-tions, and that facies variations may point to thecentre of metal-bearing volcanic-induced hy-drothermal cells. This has become a widely usedexploration criterion (Lydon, 1984; Fyon et al.,1992), although little has been documented in theliterature. The revival of gold exploration in thelast three decades added a possible sulphide faciesiron-formation to the sequence. James (1954)original facies series included black shale withdiagenetic pyrite as a possible sulphide facies; thisis in startling contrast to the interbedded chertand sulphide minerals of the observed sulphidiciron-formation. This sulphidic facies was suppos-edly gold bearing initially (Fripp, 1976) and thegold was later remobilized during deformationinto viable deposits. Work by various authors(Phillips et al., 1984) suggests that the sulphur andgold were introduced synchronous to deforma-tion, and that cherty sulphidic iron-formation isreplacement rather than a depositional facies.

While studying volcanism and hydrothermal ac-tivity in the Archean Hunter Mine group (HMG),

Fig. 1. General geology of the Abitibi (after Chown et al., 1992).

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275 265

Fig. 2. Geology of the Hunter Mine area (after Dostal and Mueller, 1997).

the authors encountered oxide and carbonate fa-cies BIF at the same stratigraphic level. Volcanicactivity was waning and there was clear evidenceof penecontemporaneous low temperature hy-drothermal activity causing subsurface replace-ment. By incorporating these observation with thegeneral literature, the following article strives todemonstrate the sometimes equivocal relation be-tween hydrothermal activity (exhalation) on thesea floor and the development of iron-formations.

2. Abitibi geology

The 2730–2726 Ma HMG volcanic rocks(Mortensen, 1987; unpublished data) constitute anintegral part of the northern volcanic zone (NVZ)in the Archean Abitibi greenstone belt (Fig. 1),

the largest coherent Archean supracrustal se-quence in the world. The study area represents thesouthern segment of the NVZ, adjacent to theE-trending Destor–Porcupine–Manneville fault,which separates the NVZ form the southern vol-canic zone (Fig. 1). The 2730–2705 Ma NVZ,composed of two distinct volcanic cycles, proba-bly accumulated in a diffuse arc setting. (Chownet al., 1992).

Volcanic cycles 1 and 2 represent incipient andmature stages of arc evolution, respectively. Vol-canic cycle 1, 1–5 km thick, is an extensive,subaqueous basalt plain with local mafic-felsic orfelsic edifices that are interstratified with or over-lain by intra-arc, volcaniclastic turbidites (Muelleret al., 1996). The volcanic edifices are submarinecomposites or stratovolcanoes generally of bimo-dal composition. These volcanic centres, dispersed

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275266

throughout the Abitibi greenstone belt, range inage from 2730 to 2720 Ma (Mortensen, 1993a,b).The 1–3 km-thick, volcanic cycle 2 (2720–2705Ma; Mortensen, 1993a,b), is preserved in thenorthern and southern extremities of the NVZ(Fig. 1) and displays, in the Chibougamau area,the evolved stages of arc development, with theemergence and unroofing of the arc (Mueller etal., 1989). Coarse clastic deposits (Mueller andDonaldson, 1992a) and contemporaneousshoshonitic volcanism (Picard and Piboule, 1986)are indicative of arc maturity. The HMG, focus ofthis study and traceable along a strike of 50 km,is part of volcanic cycle 1 in the southern segmentof the NVZ (Fig. 2).

3. Hunter Mine group

The inferred 4–5 km thick HMG (HMG; Fig.3) is a complex submarine volcanic edifice over-lain conformably by komatiites and komatiiticbasalts of the Stoughton–Roquemaure group(SRG; Eakins, 1972; Mueller et al., 1997). Thebasal to medial parts of the HMG are composedof 5–50 m thick, massive to brecciated rhyoliticlava flows and lobes and an extensive, 1.5 kmwide, felsic-dominated, dyke swarm that can betraced for 2.5 km up-section (Mueller and Don-aldson, 1992b). Collectively, these units have acalc-alkaline arc signature (Dostal and Mueller,

Fig. 3. Generalized stratigraphy of the Hunter Mine group (after Dostal and Mueller, 1997).

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275 267

Fig. 4. Outcrop map showing a portion of the upper part of the Hunter Mine group. A large enclave of oxide iron-formation occursas a rip-up within a volcanic breccia. The enclave is folded, but remained a coherent block. Note the chert jasper magnetite veinsalong synvolcanic faults (after Mueller et al., 1997).

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275268

1996). Tholeiitic, mafic dykes of SRG-affinity area minor component of the felsic-dominated dykeswarm (Dostal and Mueller, 1996). Abundant py-roclastic rocks either deposited from primaryflows or remobilized and deposited downslope viamass flow processes characterize the volcaniccomplex (Mueller and Donaldson, 1992a,b). Inaddition to pyroclastic debris, magnetite iron-for-mation and jasper-magnetite iron-formation con-stitute a minor but significant part of the earlyconstructive phase of the edifice.

The upper 500–1000 m thick transition zone ofthe HMG, directly above the dyke system (Fig. 2),is a highly variable association of pyroclastic,shale and volcaniclastic sedimentary rocks, associ-ated with iron-formation, mafic to intermediatedykes and sills and mafic and felsic lava flows

(Eakins, 1972; Mueller et al., 1997; N’dah, 1998).The mafic flow component increases towards theSRG, displaying on-lapping (Eakins, 1972) andinterstratification relationship between groupswhich is common to volcanic arc terranes(Mueller et al., 1989). Combined, iron-formation,chert beds and tuffs up to 40 m thick can betraced 5 km along strike. The summital part ofthe HMG is characterized by intense silicificationand carbonate alteration, especially evident in thevolcaniclastic rocks and shale. The presence ofsynvolcanic and synsedimentary faults parallel tothe N-trending dyke systems which served as con-duits for hydrothermal fluids, attest to significantvolcanic activity. Numerous faults are injected bydykes with dyke margins acting as vents for hy-drothermal activity (Fig. 4). Evidently, hydrother-

Fig. 5. Photographs of typical BIF. (A) Jaspilite iron-formation preserved in rip-up (Fig. 4) note white areas of recrystallization.Pencil 15 cm. (B) Typical oxide facies iron-formation (see Fig. 6) white bands recrystallized chert, black, oxide-rich layers and greygraded tuffaceous silt beds. Coin 2 cm. (C) Continuous cherty layers in tuff-turbidites with the onset of carbonate flooding (darkgrey). Pencil is 15 cm. (D) Typical carbonate facies iron-formation, disconnected layers of white recrystallized chert (notelamination) in a seemingly homogeneous matrix of siderite. Pencil is 15 cm.

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275 269

Fig. 6. Outcrop map of a stratigraphic section in the transition zone of the Hunter Mine group, showing the normal stratigraphicrelation of oxide facies iron-formation intercalated with graded volcanic tuffs (after Mueller et al., 1997).

mal activity recurred throughout the history ofthe Hunter Mine volcanic complex.

The contact of the SRG, although locally char-acterized by a zone of high strain, is depositional,and interstratification between tholeiitic mafic and

calc-alkaline felsic flows is observed at some local-ities (Fig. 4). The contact between the HMG andSRG is defined by the first appearance of komati-itic basalts. Dynamic contacts of this nature arecommon to volcanic sequences of which the

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275270

Hunter Mine and Stoughton–Roquemauregroups are no exception. The iron-formation, dis-cussed in this study, is prominent in the summitalpart of the sequence and is associated with amassive sulfide deposit (N’dah, 1998) of Mattabi-type alteration (Morton and Franklin, 1987), fel-sic lava flow breccias, and volcaniclasticsedimentary rocks.

The Hunter Mine and Stoughton–Roquemauregroups represent a south-facing homoclinal se-quence (Fig. 2) characterized by steeply dipping,W to WNW-striking strata, locally overturned tothe north. The rocks were affected by sub-green-schist facies metamorphism, characterized by aprehnite-chlorite-epidote (zoisite) assemblage. Theprefix meta, has been omitted to simplify rockdescription.

4. Sedimentary lithofacies and iron-formations

The volcaniclastic sedimentary lithofacies repre-sent an integral part of the summital sequence andmust be described in order to place the iron-for-mation deposits in a proper facies context. Thetuffs vary in thickness from 0.03 to 5 m and areintercalated with massive and brecciated felsicflows, forming continuous or lenticular units.They may be massive, or bedded and laminated.Beds of tuff 3 cm thick generally alternate withchert beds of the same thickness. Although thetuff is very fine-grained and recrystallized, thepresence of ghosts of devitrified glass shards,quartz crystals and composite quartz and matrixgrains argues in favour of a pyroclastic origin.The bedding and fine lamination indicate deposi-

Fig. 7. (A) Photomicrograph of chert layer showing fine laminations (horizontal). Random porphyroblasts and veins are composedof carbonate. (B) Photomicrograph of silicified black shale ‘cap rock’. (C) Photomicrograph of stylolitic chert being replaced byankerite porphyroblasts. (D) Photomicrograph of volcanic fragment completely replaced by siderite.

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275 271

Fig. 8. Outcrop map showing section through the transitionzone of the Hunter Mine group. Silica cap rock overlies thecarbonate alteration zone in which volcanic breccia as well asintercalated iron-formation and chert-bearing tuff are all heav-ily carbonatized. Scale bar 15m. (after N’dah, 1998)

silica may have been focused fluids passing throughthe soft sediment to deposit principally at thesediment–water interface.

5. Oxide facies iron-formation

Oxide iron-formation is widespread in the upper-most section, or transition zone of the HMG andalso occurs irregularly in sedimentary interlayersthroughout the volcanic section. It is best preservedin a large, folded rip-up clast within a volcanicbreccia (Fig. 4) where it appears as typical redbanded jaspilite iron-formation. Elsewhere in thesuccession even the low grade metamorphism hasresulted in recrystallization of the jasper.

The jaspilite iron-formation consists of delicatelybanded, brown to red, jasper layers up to 1 cmthick, alternating with thinner (0.5 cm) magnetitelayers and 1 cm thick graded silt layers (Fig. 5A).Repetition of layers is not regular. Incipient recrys-tallization along the edges of the rip-up clast resultsin conversion of the jasper to white microcrystallinequartz, but the magnetite and silt layers show novisible change.

Although this large clast displays the BIF in itspristine state, it represents less than 1

2 m of sectionat best, and the slightly recrystallized iron-forma-tion interbedded in the volcano-sedimentary stratagives a better idea of the overall depositionalenvironment. The preserved section of iron-forma-tion seldom exceeds 2 m in thickness (Fig. 6). Themore or less alternating black, dark grey and whitelayers (Fig. 5B) are composed of magnetite, gradedtuffaceous silt and chert. Although outwardly ap-pearing as magnetite-bearing iron-formation, mag-netite layers comprise only a minor portion. In thedepicted section (Fig. 5B; Fig. 6) only the lower 10cm actually contains oxide interlayers, and theremainder is composed of alternating tuffaceoussilt and chert.

In thin section, the graded beds are seen to becomposed almost exclusively of volcanic material,shards, lithic and mineral fragments, seeminglythe very fine-grained equivalents of the gradedunits higher in the section. The chert layers arecomposed of microcrystalline quartz whose con-tacts with the silt are somewhat blurred and irreg-

tion under weakly turbulent conditions, probablyresulting from the collapse of suspended sedimentclouds generated by volcanic eruptions (Lowe,1988).

Chert beds are omnipresent and always inter-spersed with finely laminated tuff. They contain fineparallel laminations of inclusions, which are com-monly carbonatized. Most are finely laminated, buta few lack internal structure. Both massive andlaminated chert are commonly brecciated, particu-larly where interlayered with iron carbonate. Somemassive chert (up to 5.5 m) forms large lenses whichreplaced many layers of tuff seen laterally in thesequence. The bedded cherts may have formed bydirect deposition, but the presence of the massivereplacement cherts suggests that the source of the

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275272

ular as a result of recrystallization. Fine lamina-tions shown by trains of minute opaque inclusionsare still faintly visible in the chert.

In summary, the oxide formation is a regularpart of the sedimentary units intercalated in theHMG. It appears to represent normal pelagicsediment deposited in periods of volcanic quies-cence. Once deposited, the sediment is ripped upand preserved within subsequent volcanic brec-cias, much as was described for turbiditic se-quences by Meyn and Palonen (1980). The fineparallel layering in the chert suggest primary de-position, or diagenesis at the sediment–water in-terface synchronous with sedimentation, possiblyfavouring the silica-rich (glassy) fine layers. Themagnetite layers may be the result of diageneticalteration of an original iron precipitate, althougha syn-sedimentary diagenesis is possible for theselayers, and might explain the irregular iron-richlayers in the primarily chert sequence.

6. Carbonate facies iron-formation

Carbonate facies iron-formation occurs over a1 km strike length of the HMG transition zoneeast of the best-preserved oxide iron-formationlocalities. It is spatially associated with evidenceof widespread hydrothermal alteration and minorsulphide mineralization of the entire transitionzone (N’dah, 1998).

Carbonate facies iron-formation is a buff andwhite coloured banded rock composed of distinc-tive cm-thick white chert layers, initially continu-ous (Fig. 5C), but most commonly slightlydiscontinuous, interbedded with medium-grainedsiderite layers up to 4 cm thick (Fig. 5D). Inextreme cases the iron formation is a chert brecciain a carbonate matrix. Some corroborative detailmay be gained from a microscopic examination ofthe iron-formation. Layers of minute opaque in-clusions within the chert (Fig. 7A), identical to

Fig. 9. Compositional analyses of carbonates from the alteration zone in the Hunter Mine group (N’dah, 1998).

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275 273

those of the oxide iron-formation witness the finelaminations largely lost in recrystallization. Sider-ite everywhere replaces chert, giving the layers asomewhat ragged appearance and resulting in thediscontinuous chert layering. Some visiblemineral grains (chiefly quartz) and ghosts ofshards visible in the carbonate indicate that muchof the carbonate replaces original tuffaceous silt.A few bedding-parallel lines of oxide grains sug-gest that original oxide layers may also be re-placed.

A typical section through the transition zone ofthe HMG (Fig. 8) shows the extent of the hy-drothermal alteration throughout the section de-scribed by N’dah (1998). The intercalated shale,BIF and volcanic breccia are almost completelyhydrothermally altered. The uppermost part ofthe section is silicified (Fig. 7B) and with associ-ated Fe-chlorite represents the first phase of hy-drothermal alteration (N’dah, 1998), whichcreated an impervious silica cap to the sequence.The second stage of alteration completely con-verted volcanic breccia and iron-formation-banded chert and silt below this cap, to car-bonate (Fig. 9). Alteration fronts preservedin the volcanic breccias show the first appearanceof random porphyroblasts of ankerite (Fig. 7C)followed by complete replacement bysiderite, commonly partly preserving the originaltexture of the rock (Fig. 7D). The tuffaceoussilt interbeds with the chert in the oxide iron-formation were easily replaced, producingthe carbonate facies iron-formation. It is impossi-ble to say if all rocks experienced an initial con-version to ankerite, or if it was formed only at theend stages of carbonatization. Sulphide mineral-ization (chiefly marcasite and pyrite) occurs prin-cipally in the altered volcanic breccia and littleoccurs in the iron-formation. The presenceof spherulitic marcasite suggests a low tempera-ture ($100°C) mineralizing fluid (Ames et al.,1993).

7. Summary

The oxide and carbonate facies iron-formationin the HMG are not the result of the same

process, nor are they necessarily the direct resultof volcanic processes. The oxide iron-formation isboth laterally widespread and common at variousstratigraphic levels in the HMG and is clearly acommon pelagic sediment at times of volcanicquiescence. The most direct relation to volcanismis clearly the quiescence. Although some massivechert is present, the very regular nature of most ofthe chert interbeds suggests primary deposition orreplacement at the sedimentary–water inter-face. The presence of quartz–jasper-bearing veinsparallel to synvolcanic faults (Fig. 4) is astrong argument for a fluid source of localvolcanic derivation, although their precise timingwith respect to the BIF is not clear. Theregular interaction between hot seawater andglass-rich, chemically-active volcanic breccias andsediments can hardly be ruled out as the ultimatesource for the Si and Fe. It is a fact, however, asis shown dramatically in metallogenic maps ofsedimentary rock belts northeast of thestudy area, that there is a strong correlation be-tween thick sedimentary sequences and the occur-rence of BIF in the Abitibi belt, and the oxideBIF may well be the end result of the action ofcold seeps filtering up through the fertile vol-canogenic sediments to the sea floor (cf. sub-marine fans, Fyon et al., 1992). They may also bethe result of a hydrothermal system not related tosulphide deposition as described by Leistel et al.(1998)

The carbonate facies iron-formation in theHMG is clearly the result of in situ low tempera-ture hydrothermal replacement which formed asilica cap and then replaced the volcaniclasticrocks and the chert tuff formation below the capwith siderite as the hydrothermal system evolved.The hydrothermal activity is probably the resultof late-stage volcanic activity and seawater circu-lation, but the BIF itself is a replacement depositand not the result of direct seafloor precipitationinduced by volcanism. A similar system is docu-mented in the Iberian Sulphide belt (Leistel et al.,1998) where Fe–Mn carbonates replace chert in-terbeds, and the Helen Formation, long consid-ered the type volcanic exhalative iron-formationformed in a similar manner (Morton and Nebel,1984).

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275274

8. Conclusion

The two iron-formations in the HMG are spa-tially, but not temporally related. Oxide faciesiron-formation formed first at the seawater inter-face, partly replacing the finer-grained portions ofthe turbiditic tuffs. Subsequently, a low-tempera-ture hydrothermal system developed, which firstformed a silica cap rock, and, as the hydrothermalsolution evolved, replaced the underlying volcani-clastic sequence with siderite. The interbeddedchert-tuff sequence became interbedded chert andsiderite,- an iron-formation. This example of oxideand carbonate facies iron-formation demonstratesthe difficulty of inferring a simplistic one-to-onecorrelation between iron-formations and volcanicactivity. The oxide iron-formation appears to be anormal pelagic deposit in a volcano-sedimentaryenvironment. It may be the result of direct volcanicexhalation, but is more likely produced from ex-halative fluids derived from rock-water interaction.As demonstrated in this study, it may occur withinthe central volcanic complex, rather than being adistal manifestation. The carbonate iron-forma-tion, on the other hand, is of replacement originand cannot be used as an indicator of sedimentaryfacies, although its obvious relation to a hydrother-mal cell makes it alone a promising prospectingtool. These results show that the broad interpreta-tion of banded iron-formation as a distal equivalentof volcanogenic massive sulphide deposits isunwarranted.

Acknowledgements

The authors are grateful to Exploration NorandaLtee. for logistical support. Additional support forfield and analytical work from NSERC grants toEHC and WUM is also acknowledged. E.N.’s MSc.study was made possible by a scholarship from thegovernment of the Cote d’Ivoire. The authorsthank K. Caron for assistance with the field work,C. Dallaire with the drafting and R. Daigneault forgeneral advice and encouragement. Critical readingby M. Kimberley and P. Thurston greatly improvedthe manuscript. Lithoprobe Contribution No.1128.

References

Ames, D.E., Franklin, J.M., Hannington, M.D., 1993. Miner-alogy and geochemistry of active and inactive chimneysand massive sulfide, middle valley, northern Juan de FucaRidge: an evolving hydrothermal system. Can. Min. 31,997–1024.

Chown, E.H., Daigneault, R., Mueller, W., Mortensen, J.,1992. Tectonic evolution of the northern volcanic zone,Abitibi belt, Quebec. Can. J. Earth Sci. 29, 2211–2225.

Dimroth, E., Chauvel, J.J., 1973. Petrography of the Sokomaniron formation in part of the central Labrador Trough,Quebec. Can. Geol. Soc. Am. Bull. 84, 111–134.

Dimroth, E., 1986. Depositional environments and tectonicsettings of the cherty iron-formations of the CanadianShield. Geol. Soc. India 28, 239–250.

Dostal, J., Mueller, W., 1996. An Archean oceanic felsic dykeswarm in a nascent arc: the Hunter Mine group, Abitibigreenstone belt. J. Volcanol. Geotherm. Res. 72, 37–57.

Dostal, J., Mueller, W., 1997. Komatiite-flooding of a riftedrhyolitic arc complex: geochemical signature and tectonicimplications, Stoughton–Roquemaure group, Abitibigreenstone belt. Can. J. Geol. 105, 545–563.

Eakins, P.R., 1972. Roquemaure township: Department Natu-ral Resources, Quebec Geol. Rep. 150, 69p.

Fripp, R.P., 1976. Stratabound gold deposits in Archeanbanded iron-formations, Rhodesia. Econ. Geol. 71, 58–75.

Fyon, J.A., Breaks, F.W., Heather, K.B., Jackson, S.L., Muir,T.L., Stott, G.M., Thurston, P.C., 1992. The metallogenyof metallic mineral deposits in the Superior Province ofOntario. In: Thurston, P.C., Williams, H.R., Sutcliffe,R.H., Stott, G.M. (Eds.), Geology of Ontario. OntarioGeological Survey, pp. 1091–1174.

Gross, G.A., 1980. A classification of iron formations basedon depositional environments. Can. Miner. 18, 171–187.

Holland, H.H., 1984. The chemical evolution of the atmo-sphere and oceans. Princeton University Press, Princeton,NJ, p. 582.

Kimberley, M.M., 1979. Geochemical distinctions among envi-ronmental types of iron formations. Chem. Geol. 25, 185–212.

Kimberley, M.M., 1989. Exhalative origins of iron formations.Ore Geol. Rev. 5, 13–145.

Kimberley, M.M., 1994. Debate about ironstone: has solutesupply been surficial weathering, hydrothermal convectionor exhalation of deep fluids? Terra Nova 6, 116–132.

Leistel, J.M., Marcoux, E., Deschamps, Y., 1998. Chert in theIberian pyrite belt. Min. Depos. 33, 59–81.

Lowe, D.R., 1988. Suspended-load fallout rate as an indepen-dent variable in the analysis of current structures. Sedimen-tology 35, 765–776.

Lydon, J.W., 1984. Volcanogenic massive sulphide deposits. 1.A descriptive model. In: Roberts, R.G., Sheahan, P.A.(Eds.), Ore Deposit Models, vol. 3. Geoscience CanadaReprint Series, pp. 145–153.

Meyn, H.D., Palonen, P.A., 1980. Stratigraphy of an Archeansubmarine fan. Precambrian Res. 12, 257–285.

E.H. Chown et al. / Precambrian Research 101 (2000) 263–275 275

Mortensen, J.K., 1987. U-Pb zircon ages for volcanic andplutonic rocks of the Noranda-Lac Abitibi area, AbitibiSubprovince, Quebec. Curr. Res. A Geol. Surv. Can. 87(1A), 581–590.

Mortensen, J.K., 1993a. U-Pb geochronology of the easternAbitibi Subprovince. 1. Chibougamau-Matagami-Joutel.Can. J. Earth Sci. 30, 11–28.

Mortensen, J.K., 1993b. U-Pb geochronology of the easternAbitibi Subprovince. 2. Noranda-Kirkland lake area:. Can.J. Earth Sci. 30, 29–41.

Morton, R.L., Nebel, M.L., 1984. Hydrothermal alteration offelsic volcanic rocks at the Helen Siderite deposit, Wawa.Ont. Econ. Geol. 70, 1319–1333.

Morton, R.L., Franklin, J.M., 1987. Two-fold classification ofArchean volcanic-associated massive sulfide deposits.Econ. Geol. 82, 1057–1063.

Mueller, W., Chown, E.H., Sharma, K.N.M., Tait, L., Roche-leau, M., 1989. Paleogeographic evolution of a basement-controlled Archean supracrustal sequence,Chibougamau-Caopatina, Quebec. J. Geol. 97, 399–420.

Mueller, W., Donaldson, J.A., 1992a. Development of sedi-mentary basins in the Abitibi belt: an overview. Can. J.Earth Sci. 29, 2249–2265.

Mueller, W., Donaldson, J.A., 1992b. A felsic feeder dykeswarm formed under the sea: the Archean Hunter Mine

group, south-central Abitibi belt, Quebec, Canada. Bull.Volcan. 54, 117–134.

Mueller, W.U., Daigneault, R., Mortensen, J.K., Chown,E.H., 1996. Archean terrane docking: upper crust collisiontectonics, Abitibi greenstone belt, Quebec, Canada. Tec-tonophysics 265, 127–150.

Mueller, W., Daigneault, R., Chown, E.H., 1997. Archeanterrane docking along the Destor-Porcupine-Mannevillefault: field trip A-7 Guidebook. GAC/MAC, Ottawa,Canada, p. 68.

N’dah E. 1998. Interpretation des textures des sulfures et del’alteration, d’une systeme hydrothermale dans la GroupHunter mine, MSc thesis. University Quebec a Chicoutimi,Canada.

Picard, C., Piboule, M., 1986. Petrologie des roches vol-caniques du sillon de roches vertes archeennes deMatagami-Chibougamau a l’ouest de Chapais (Abitibi est,Quebec). 1. Le Groupe hautement potassique d’Opemisca.Can. J. Earth Sci. 23, 1169–1184.

Phillips, G.N., Groves, D.I., Martyn, J., 1984. An epigeneticorigin for Archean banded iron formation-hosted golddeposits. Econ. Geol. 79, 162–171.

Simonson, B.M., 1985. Sedimentological constraints on theorigin of Precambrian iron-formation. Geol. Soc. Am.Bull. 96, 244–252.

.